How Do You Spot a Sub Bid That Is Too Low?

Some low bids are simply good prices: a sub with crews coming free, a better supplier deal, a more efficient means and method. Others are mistakes, and a mistake found after award becomes a change order fight, a default, or a sub who stops showing up. The job on bid day is to tell the two apart before you commit. Do it on leveled numbers: a base bid that excludes a scope item is low for an obvious reason, and leveling has already priced it with a plug.

Public owners have a written habit worth copying. After bids are opened, federal contracting officers "shall examine all bids for mistakes", and where they have reason to believe a mistake may have been made, "shall request from the bidder a verification of the bid, calling attention to the suspected mistake", before award (FAR 14.407-1). The Federal Highway Administration's bid review guidelines list factors such as the comparison of the bids against the engineer's estimate, the number of bids submitted and the distribution or range of bids received (FHWA). Neither binds a private GC, but both describe the right sequence: compare, suspect, ask, then decide.

A single percent cutoff is a prompt, not a rule. A bid 10 percent under the median on a trade where bids usually cluster tightly may deserve more attention than one 20 percent under on a trade where they always scatter. Read the breakouts before you read anything into the percentage.

Five bidders priced the plumbing package on a fictional two-level clinic. The GC's estimate for the trade is $640,000. After leveling for excluded scope, the totals are below.

With five bids the median is the third: $648,500. Pinecrest is $151,500 under it ($151,500 ÷ $648,500 = 23 percent) and $115,000 under the next bidder. The other four sit within 6 percent of the median, so the field is not low; one bidder is. The breakouts point at sanitary and storm piping: $96,000 against $184,000 to $204,000 from everyone else. Pinecrest's quantity sheet lists 1,150 linear feet of underground piping; the drawings show about 2,300, both levels. It priced one level.

Its bid also acknowledges Addendum 1 but not Addendum 2, which added two grease interceptors. The four bidders who priced Addendum 2 carried $14,500 to $16,000 for them. Corrected on the GC's figures, Pinecrest would be $497,000 + $96,000 (the second level at the same price) + $15,500 (the addendum) = $608,500: still the lowest, by $3,500 under Keystone's $612,000. That is why the next step is to ask, not to throw the bid out.

1. The bidder confirms its price as submitted. Get the confirmation in writing, including the scope you questioned, then weigh the risk: its backlog, its bonding, how it has performed for you. A confirmed low price on fully covered scope is a good price.
2. The bidder shows a mistake and the price it intended. Federal rules allow correction when "clear and convincing evidence establishes both the existence of the mistake and the bid actually intended" (FAR 14.407-3). The same test is a fair one privately: a corrected bid backed by worksheets, compared with the others on the same basis.
3. The bidder shows a mistake but cannot show what it intended, or asks to withdraw. Let it withdraw. The federal rule allows withdrawal on evidence that is clear as to the mistake even when it is not clear as to the intended bid. A sub forced to perform at a price it knows is wrong is the risk you were trying to avoid.
4. The bidder does not answer. Treat its number as unverified, and do not carry it into your estimate or award on it.

Two things not to do. Do not tell the low bidder what the others bid or by how much it is low: that is bid shopping, and it damages your coverage on every future job. And do not plug a low bidder's mistake silently and award at its original number, expecting to argue scope later. Write down what was asked, what came back and why you awarded as you did.

In Conflo, bid leveling is a tab of each trade bid, seen only by the general contractor's team. Each submitted bid is a column and each scope item a row, showing the bidder's breakout amount where it gave one; a scope item a bidder excluded is marked as missing scope until you price it with a private adjustment. Each bid's leveled total is its base bid plus your adjustments, and bids sort from low to high.

Under each leveled total the grid shows how far it sits from the median, as a percent above or below, and the header shows the median and the spread across all bids, so a bid 23 percent under the rest is plain at a glance. Exclusions and qualifications sit on their own rows beside each bid, and a bid missing a trade addendum is flagged on the Bidders tab; selecting it in leveling brings up a warning to confirm past. When a bidder revises its bid, the column shows the revision number.

Spotting a bid that is too low is leveling plus curiosity: compare the leveled totals with the median and your estimate, find the item where the gap sits, check quantities and addenda, and ask the bidder to verify in writing.

The answer decides the award. A confirmed price is a good price; a real mistake is corrected or withdrawn before you commit, not discovered by your superintendent in the third month.

See how bid leveling works in Conflo

1. FAR 14.407-1, General (Mistakes in bids) — Acquisition.gov, U.S. General Services Administration. Read October 22, 2026.
2. FAR 14.407-3, Other mistakes disclosed before award — Acquisition.gov, U.S. General Services Administration. Read October 22, 2026.
3. Guidelines on Preparing Engineer's Estimate, Bid Reviews and Evaluation (October 7, 2021), Bid Review Factors — Federal Highway Administration. Read October 22, 2026.